[BOUNTY CLOSED] Open Source (CC) Paper Wallet Kit for safe offline coin storage

[75RTUGA]

My work...

[1714121829]

1714121829

[canton]

Aantonop,

Any tests/results yet to see whether or not the scratch-off seal on heavy stock is susceptible to "candling", or if they help with waterproofing? I'm especially curious to know if these stickers are an effective light-guard.

[batcoin]

Have you experimented to make sure you can't use a bright laser pen or other high intensity light source to transmit enough light through the scratch-off material to see (scan) the QR code on the other side?

I've been working on a folded-style paper wallet and someone suggested this might be a problem. I was skeptical, but after some experiments I was absolutely able to reveal some details that I believed would be obfuscated behind multiple layers of paper & ink.



The thing I'm especially interested in re: the scratch-off stickers is how they might protect against water damage when the keys are printed using an inkjet printer. Have you run some tests in that regard, seeing how a scratch-off covered inkjet-printed QR code survives when wet?

Here are the fairly horrifying results from my own experiments to this end:

Have there been any ideas on dealing with these issues? Is it possible to have the printing company laminate a plastic printing medium for the private keys that would then be waterproofed once the scratch-offs are applied? I am thinking private key sandwiched between two plastic layers.
If it is possible to have a waterproof/resistant printing laminate applied, could there be a black and white security pattern behind it to help prevent candling? I suppose this would require a little bit of tweaking to the original design...

[superbit]

Why not just purchase waterproof tear proof paper and do it on that?

[acorn]

The paper being provided should be fine, but I think the concern is the ink being used by the consumer once they receive their prepared sheets. The user prints QR codes on to the paper at home. You _could_ warn users to use laser printers where possible but I guess it comes down to how much of a ready-out-of-the-box solution you want this to be.

[batcoin]

Why not just purchase waterproof tear proof paper and do it on that?

The paper being provided should be fine, but I think the concern is the ink being used by the consumer once they receive their prepared sheets. The user prints QR codes on to the paper at home. You _could_ warn users to use laser printers where possible but I guess it comes down to how much of a ready-out-of-the-box solution you want this to be.

Right. And how many people have laser printers now? I think there is a large number that still use inkjets, no? I am concerned about the inkjet ink running if it gets wet. I think that is a solution worth baking into the design from the get-go. We are talking about people using this to safely store their life savings after all.

[acorn]

Right. And how many people have laser printers now? I think there is a large number that still use inkjets, no? I am concerned about the inkjet ink running if it gets wet. I think that is a solution worth baking into the design from the get-go. We are talking about people using this to safely store their life savings after all.

I absolutely agree with you. As far as what can be done about the paper itself, I think there is very little we (the kit providers) can do. Some paper may absorb water a little less ('tough paper') but inkjet printer inks are all water soluble and, when wet enough, the ink will definitely run. If the front side (printed side) is covered with a clear plastic from the scratch off sticker, I wonder if sealing the back with a plain clear plastic sticker (not scratch-off) would help to minimize absorption close to the printed QR?

If the paper itself were water-resistant or hydrophobic, the printed qr codes would take a long time to dry or might not settle into the paper at all, so I have a feeling covering the area front and back with some sort of plastic film after printing the codes would be the only way.

Thoughts?

I remember in art school we used to spray our work with hairspray and it helped prevent smudging and deterred moisture, but I'm not sure how helpful that is lol

[aantonop]

Actually, I think Scotchguard would work, but it cannot be flat-packed in an envelope. We could add it to a "tips" section

I will run some light and water tests today. Please standby. Sorry for the delay, but my other project (OpenBitcoinStore) became huge overnight... (see https://bitcointalk.org/index.php?topic=161264.0;all if interested)



REMINDER: Submit designs by end of week.

I will be copying all the submitted designs to a single location and adding a voting widget, so everyone can vote on the best design.

Thanks!

[batcoin]

...I wonder if sealing the back with a plain clear plastic sticker (not scratch-off) would help to minimize absorption close to the printed QR?

If the paper itself were water-resistant or hydrophobic, the printed qr codes would take a long time to dry or might not settle into the paper at all, so I have a feeling covering the area front and back with some sort of plastic film after printing the codes would be the only way.

Thoughts?

I remember in art school we used to spray our work with hairspray and it helped prevent smudging and deterred moisture, but I'm not sure how helpful that is lol

I may not have worded it well, but this is exactly what I was trying to say. I am almost certain there is some kind of plastic film that inkjets could print on. I think that if that plastic could be pre-applied to the private key field of the wallets, people could still print on them and then just apply the plastic/scratch-off covers. I will try to work up a diagram if anyone wants.

[aantonop]

I think I know what you're talking about.

3M I think sells a coating like that. I should also ask the printing company. This could really add value to the product.

[batcoin]

I slapped this together really quick and dirty.



So we have the design with the security pattern printed on the wallet. They printing companies can hopefully apply the waterproof printing label over that security pattern. Something like this, but probably opaque white to help with future scanning?

The end users can slide these in their inkjets and print away. Then they apply the scratch-off adhesive, which would also add some protection to the private keys.

[aantonop]

I slapped this together really quick and dirty.



So we have the design with the security pattern printed on the wallet. They printing companies can hopefully apply the waterproof printing label over that security pattern. Something like this, but probably opaque white to help with future scanning?

The end users can slide these in their inkjets and print away. Then they apply the scratch-off adhesive, which would also add some protection to the private keys.

Our beloved template designer Acorn, has also made a background security pattern that is very cool, using the BTC symbol repeated in different angles. I was planning on printing that on the back side of the paper behind of each QR code, to make it impossible to shine-through.

We could also do it on the front, it would require more stickers.

Let's do more research... this is an exciting idea and development. Cantor brought us a completely new "user problem" which actually can become one of the unique features and benefits of this project.

[batcoin]

I slapped this together really quick and dirty.



So we have the design with the security pattern printed on the wallet. They printing companies can hopefully apply the waterproof printing label over that security pattern. Something like this, but probably opaque white to help with future scanning?

The end users can slide these in their inkjets and print away. Then they apply the scratch-off adhesive, which would also add some protection to the private keys.

Our beloved template designer Acorn, has also made a background security pattern that is very cool, using the BTC symbol repeated in different angles. I was planning on printing that on the back side of the paper behind of each QR code, to make it impossible to shine-through.

We could also do it on the front, it would require more stickers.

Let's do more research... this is an exciting idea and development. Cantor brought us a completely new "user problem" which actually can become one of the unique features and benefits of this project.

Yeah, I was just putting it there for the sake of demonstration. My graphic design skills are lack-luster to say the least. I think Acorn's security background design looks pretty nice. I also don't know much about how well security features work. I just thought these couple ideas might be helpful solutions to the problems that canton brought up.

[aantonop]

Yeah, I was just putting it there for the sake of demonstration. My graphic design skills are lack-luster to say the least. I think Acorn's security background design looks pretty nice. I also don't know much about how well security features work. I just thought these couple ideas might be helpful solutions to the problems that canton brought up.

I did not mean to sound dismissive, quite the opposite. I think your idea of printing it in front and covering with a printable coating is very good and could be less expensive than the back-printed idea. Your design is fine.

I was aiming to encourage more discussion and research on this. Apologies if it came off sounding otherwise!

[batcoin]

Yeah, I was just putting it there for the sake of demonstration. My graphic design skills are lack-luster to say the least. I think Acorn's security background design looks pretty nice. I also don't know much about how well security features work. I just thought these couple ideas might be helpful solutions to the problems that canton brought up.

I did not mean to sound dismissive, quite the opposite. I think your idea of printing it in front and covering with a printable coating is very good and could be less expensive than the back-printed idea. Your design is fine.

I was aiming to encourage more discussion and research on this. Apologies if it came off sounding otherwise!

It's all good. I didn't take it like that. If the Bitcoin security pattern on the back of the wallet would be sufficient, then I would vote for that. I think it looks a lot nicer and I think people would find it more attractive than simply having a blank back to the note.

[aantonop]

The great light,  liquids and transparency experiment (v.2.0) has begun.

This is all on regular 80mg bright white color laser/copier paper. Cheap and thin. Not the kind of paper we will use for the print production, but the kind most users have at home.

Here's the first set of photos, as a high-res album:

http://imgur.com/a/KOOH6#0


Light Test


The photos were taken from a very close distance, with the notes pinned to a lightbox (CFD light).

As you can see, the security sticker is absolutely opaque, no possible way to see through.












The comparison to the sharpie (black marker pen) is quite interesting. In normal light and with the naked eye, the sharpie was obscuring the QR code and was opaque, no chance of seeing the QR. On the lightbox, the difference in color and shading is so big that the sharpie ink seems almost transparent.










Now looking at the reverse side, we see that the inkjet printer's ink has saturated the paper, and bled through to the other side. Even though the security sticker does not let light through, the ink is reflected from the camera flash, through the paper. With very little image re-touching it would be scannable.










On the laser, there is no visible code.












Conclusion of light test: My opinion on these results is that with a security pattern on the back, a sticker on the QR and thicker paper, we will be able to solve the compromise-with-light problem. The security stickers work exactly as advertised and are light-proof.

Liquid Test


The test involved adding 2 cc of liquid (applied with a syringe, for volume consistency), on each note.

I tested notes with and without the security stickers.
I tested three liquids: 2 cc Water, 2 cc vinegar (acid), 4 cc sports drink (red color dye)
I tested three arrangements: Liquid on the front, liquid on the back, liquid on the front and folded

Here you can see the test-grid, about a minute after I finished applying the liquids from left to right. The left side is completely saturated, the right side is beginning to soak.







Finally, I have kept several sets of "control" notes with and without stickers. Tomorrow I will add water to some, to see if there is a difference in the result if the ink has had time to be fully absorbed, or the sticker has had time to fully bond with the paper.

This will result in 42 combinations (30 now, 12 more tomorrow). I will photograph each note, scan the QRs (or try to) and tabulate and analyze the results, in a couple of days.

Let's see how the various combinations hold up!

[casascius]

Something to keep in mind with respect to water-damaged QR codes:

The human eyes are far better at reading QR codes than virtually any QR code scanner.  The only difference is the human brain lacks the capacity to decode the meaning of the QR code.

If ever you encounter an unreadable QR code, but you can make out which squares are black and white, even if you have to guess on some of them, then you can recover the QR code simply by recreating it manually from a paint program or even a piece of graph paper, and then scanning the result.

QR codes contain redundancy, so they are tolerant of some of the squares being the wrong color, or missing.  If part of the code is hard to read, a good guess never hurts.  The only thing that is critical, is that QR won't be tolerant of problems with alignment, or skipping rows, or losing track of which row/column you're on - the sanctity of the grid, and positioning of things relative to the grid, are critical.

[aantonop]

Preliminary liquids report:

Worst damage is done on water soaked, inkjet printed, no sticker, folded over. The background colors bleed into the QR black ink and become a soup.

Least damage is on the water-soaked, laser printed, sticker covered, unfolded notes. They seem completely unaltered (haven't scratched the sticker off yet).

The biggest problem seems to be the color ink from inkjets, especially if two printed faces are touching and get wet (either double-side printing on inkjet or folded notes).

I think that the combination of pre-printed backgrounds, thicker paper, and stickers will eliminate the water problem almost entirely. If the users can use laser instead of inkjet for the QR's the problem drops to zero.

I have applied water to the another set of notes, now the ink has had time to dry, to see if it makes a difference.

Tomorrow I will properly tabulate the results. At this point I think we have about a 5% failure rate (QR is likely unreadable, or difficult to scan) across the entire sample.

[superbit]

Again, use better paper.  I use this stuff all the time, it will take inkjet fine and you can then dump water on it and it won't smear.

http://www.waterproofpaper.com/inkjet-paper.shtml

[aantonop]

ONE DAY to go.

As I promised in the beginning of this project, the choice of the bounty design will be open to the community at large. I will advertise for the vote here and on reddit.

I will be using an online voting service to handle the public vote. You can see the design submissions here:

http://tricider.com/brainstorming/poxx

If you are a designer and want to participate in the vote, please submit your design by going to the link above, "Add Idea", upload an image and description with your name/nickname.

By submitting your design you agree for it to be released under CC-BY-SA, upon payment of the bounty. Any designs not paid will remain the property of the designers, though I would encourage sharing regardless.

Submissions must be complete and final before end of day Sunday. If it still says "Sunday" in whatever timezone, you can submit.

On Monday I will post the link more broadly and ask for votes. When the vote ends, the winning design gets 2.5BTC. The second design gets 2.5 BTC. I pick one for production and this bounty is over.

I reserve the right to revise, extend or cancel the vote if I see manipulation, double voting or other problems. I reserve the right to select a different design for the print production run that I will be paying for, though it is highly unlikely.

Thank you for your participation and amazing support for this project.

Any comments, suggestions or objections, you have ONE day to raise them before this policy is considered final.

Andreas